Neuropsychiatric disorders and COVID-19

Maxime Taquet and colleagues reported an increased incidence of neurological and psychiatric disorders in patients diagnosed with COVID-19 (ie, group 1) compared with two matched control cohorts: patients diagnosed with influenza (ie, group 2) and patients diagnosed with any respiratory tract infection, including influenza (ie, group 3).¹ In my opinion, having two control groups containing patients with influenza is a shortcoming of the study, and patients with influenza in group 3 should have been transferred to group 2. The authors suggested that the potential mechanisms for the association of neurological and psychiatric disorders with COVID-19 include viral invasion of the CNS, hypercoagulable states, neural effects of the immune response, and psychological and other implications of a COVID-19 diagnosis for people with common psychiatric disorders (eg, mood and anxiety disorders).¹ Other mechanisms, however, should be considered. One mechanism is that morbidity could have increased if patients did not attend necessary medical appointments during the COVID-19 pandemic because of decreased access to medical services during lockdown and semi-lockdown periods or patients’ anxiety at getting infected.

The authors did not report information about the anti-COVID-19 medications that were given to the patients in group 1 during or after hospitalisation. Neurotoxic drugs that are used to treat COVID-19 include daptomycin, linezolid, lopinavir, ritonavir, hydroxychloroquine, cisatracurium, clindamycin, tocilizumab, and glucocorticoids.² Neuropathy or myopathy in patients with COVID-19 requiring treatment in intensive care units can also result from bedding (ie, compression neuropathy), compartment syndrome, artificial nutrition, infection, electrolyte disorder, or sepsis (ie, critically ill neuropathy or myopathy).² Anti-COVID-19 drugs can also be myotoxic, for example, chloroquine is associated with myopathy and myasthenia.³

Taquet and colleagues distinguished Guillain-Barre syndrome and nerve root disorders on the basis of ICD codes in the database. However, the ICD system is incomplete: Guillain-Barre syndrome is a classic nerve root disorder and some doctors might encode Guillain-Barre syndrome under nerve root disorders, whereas other doctors might use the ICD code for Guillain-Barre syndrome. Therefore, these two groups should be assessed together.

The authors excluded patients with a diagnosis of COVID-19 or a positive test for SARS-CoV-2 from the control cohorts. However, it is not known whether all patients in the control groups were actively tested for SARS-CoV-2 infection: patients from the control groups need to be SARS-CoV-2 negative to serve as controls, which can be ensured only by systematic PCR tests in each patient.

Major outcomes that were not considered in the evaluation included cerebral vasculitis, venous sinus thrombosis, seizures or epilepsy, cranial nerve affection, myelitis, acute disseminated encephalomyelitis, and headache.⁴ Particularly, venous sinus thrombosis should have been included in the evaluation as COVID-19 is associated with an increased risk of thrombosis.⁵ The authors also did not consider hyperlipidaemia or atherosclerosis as pre-existing comorbidities: these might worsen during COVID-19 and might contribute to the increased incidence of neuropsychiatric disease.

I declare no competing interests.